Inspiration for Study Composers III

If you have an iOS device (iPad or iPhone), browsing the PDF file will not work properly,

The following PDF file with 16 fantastic practical endgames was send by Hergert, Volker 1963.

Also the endgames can be replayed below.

hopefully it may inspire composers to create new Endgame studies !

[Event "Simmerath Club Championship"] [Site ""] [Date "1983"] [Round ""] [White "Theo Klimas"] [Black "Volker Hergert"] [Result "0-1"] [Annotator "Hergert,Volker"] [SetUp "1"] [FEN "1r5R/8/3p4/3k2P1/8/P7/1rp3PP/R1K5 b - - 0 35"] [PlyCount "11"] [SourceVersionDate "2026.02.24"] 
 {In a completely lost position, White had just played 35.Rh3-h8 to give me a final and hard nut to crack.} 35... R8b7 $1 {The only winning move! Obviously, 35…Rb1+ 36.Kxc2 R8b2+ 37.Kc3 Rb3+ 38.Kc2 R3b2+ is only a perpetual. Also, 35...R8b6? fails to do the job due to 36.g6 Rb1+ 37.Kxc2 Rxa1 38.g7 and Black has nothing better than to escape into another perpetual: 38...Ra2+ 39.Kc1 Ra1+ (=).} (35... Rxh8 $2 36. Kxb2 $11) 36. Rh7 {36.g6 no longer works as after 36...Rb1+ 37.Kxc2 Rxa1 White can’t push his g-pawn. So, the idea of 35…R8b7 lies in the control of the g7 square.} R7b6 $1 {And now this is in fact the only winning move, e.g. 36...R7b3? would move the rook too close to White´s king... 37.a4 Rb1+ 38.Kxc2 = or 36...R7b5? gets into the reach of the a-pawn... 37.a4 R5b4 38.Rh4 Rb6 39.a5 Rb1+ (39...R6b5? 40.Rha4 +–) 40.Kxc2 R6b2+ 41.Kc3 Rb3+ (=).} 37. g6 Rb1+ 38. Kxc2 Rxa1 39. g7 Ra2+ 40. Kc1 Rxg2 $19 {0-1 This is the difference to playing 35…R8b6: the white rook is now on h7 instead of h8, thus White can’t promote the pawn this time.} 0-1

[Event "Aachen"] [Site ""] [Date "1983"] [Round ""] [White "Hergert"] [Black "Schmitt"] [Result "1-0"] [Annotator "Hergert,Volker"] [SetUp "1"] [FEN "8/pp6/8/1PP5/P4k2/5Prp/3K1R2/8 w - - 0 39"] [PlyCount "31"] [SourceVersionDate "2026.02.24"] 
 39. Kd3 $1 {The only move that keeps the balance. The king intends to support the pawn majority on the queenside. Instead, the obvious alternatives 39.Ke2? and 39.Rf1? would lose to 39.Ke2 Rg1 -+ or 39…Rg5 -+ and 39.Rf1 h2 40.Rh1 Rh3 41.c6 bxc6 42.bxc6 Kg3 43.c7 Rh8 -+. The rook can counter any pawn break-through on the queenside by moving behind the white pawn on the first rank. The nice aspect of this key move is that it is not only forced to save the game, but that it also contains a nasty trap.} Rxf3+ $2 {Loses by force. With any other serious candidate move, Black could have achieved a draw.} 40. Ke2 $1 Rxf2+ 41. Kxf2 Ke5 42. a5 Kd5 (42... Ke6 43. c6 b6 (43... bxc6 44. b6 axb6 45. a6 $18) 44. axb6 axb6 45. Kg3 $18) 43. c6 Kd6 (43... bxc6 44. b6 axb6 45. a6 $18) 44. cxb7 Kc7 45. a6 Kb8 46. Kg3 Kc7 47. Kxh3 Kb8 48. Kg4 Kc7 49. Kf5 Kb8 50. Ke6 Kc7 51. Kd5 Kb8 52. b6 $1 axb6 53. Kc6 Ka7 54. Kc7 $18 {1-0} 1-0

[Event "Aachen"] [Site ""] [Date "1984"] [Round ""] [White "Volker Hergert"] [Black "Vladimir Budde (ELO 2360)"] [Result "1/2-1/2"] [Annotator "Hergert,Volker"] [SetUp "1"] [FEN "8/3k4/r2n1p2/8/6PK/7P/8/3R4 b - - 0 1"] [PlyCount "38"] [SourceVersionDate "2026.02.24"] 
 1... Ra5 $2 {Black wants to take control over the fifth rank to prevent White from either moving his king to h5 or to push his g-pawn. But for tactical reasons the move is a mistake. Instead, Black could have won with precise play. That said, the winning move orders are not easy to detect over the board.} (1... Ke7 {easiest} 2. Kh5 (2. g5 f5 $1 $19) 2... Ra5+ 3. Kg6 (3. Kh6 Nf7+ 4. Kg7 (4. Kh7 Ng5+ 5. Kh6 Nxh3 $19) 4... Rg5+ 5. Kh7 Kf8 $19) 3... Rg5+ 4. Kh7 Nc8 $1 5. Rc1 Nb6 $1 6. Rc7+ Kf8 $1 7. Rb7 Nd5 $19) (1... Ke6 2. Kh5 Kf7 $1 (2... Ra5+ $2 3. Kh6 Nf7+ 4. Kg7 Rg5+ 5. Kf8 Ne5 6. Re1 $11) 3. Rc1 Ra3 $1 4. h4 Ra4 $1 5. Rc7+ Ke6 6. Rc1 Nc4 $1 7. g5 f5 8. Kh6 Ne5 9. h5 f4 10. g6 Kf5 11. g7 Ra8 12. Kh7 Nd7 $19) 2. g5 $1 {I could not calculate the consequences of 2...f5. However, what I saw was that I would reach an easy draw if Black captures on g5 and that I would for sure lose without the push of the g-pawn. So, I gave it a try.} f5 (2... fxg5+ 3. Kh5 (3. Kg4 Ke6 4. h4 $11) 3... Ke6 4. h4 $11) (2... Rxg5 3. Rxd6+ $1 Kxd6 {stalemate}) 3. g6 $1 (3. Kh5 $2 Ke6 4. g6 Ra3 5. h4 f4 6. Kg5 Ke5 $19) 3... Ra2 $1 {The most challenging continuation. Black wants to answer 4.g7 with 4...Rg2. It’s an easy draw after 3...Ke6 4.g7= or 3…Ra8 4.Kg5/Kh5 =.} 4. Rg1 {Now, 5.g7 is a real threat. Instead, 4.g7? Rg2 –+ and 4.Kg5? Ke6 -+ lose.} Ra8 5. g7 $1 (5. Kg5 $2 {fails in a study-like way:} Ke6 6. g7 Kf7 7. Rc1 (7. Kh6 Ne4 $19) 7... Kxg7 8. Rc7+ Kf8 $1 9. h4 Ra4 $1 10. Rd7 (10. h5 Rg4+ 11. Kf6 f4 12. Rd7 f3 $1 13. Rxd6 Rf4+ 14. Kg5 f2 $1 $19) 10... Rd4 11. Ra7 Nc8 $1 12. Ra8 (12. Ra6 Rd6 $19) 12... Rd8 13. h5 Kf7 $19) 5... Rg8 6. Rg6 $1 (6. Kh5 $2 {is refuted by transition to a pure pawn endgame:} Ne8 7. Kh6 Rxg7 8. Rxg7+ Nxg7 9. Kxg7 f4 10. h4 f3 11. h5 f2 12. h6 f1=Q 13. h7 Ke6 14. h8=Q {and the winning technique in this setup is well-known:} Qg2+ 15. Kf8 Qf3+ 16. Kg7 Qg4+ 17. Kf8 Qf5+ 18. Kg7 Qg5+ 19. Kh7 Kf7 $19 {The move 6.Rg6 prepares 7.Kg5.}) 6... Ne4 {Prevents 7.Kg5, which is necessary if Black wants to win, e.g. 6...Ne8 7.Kg5 =.} 7. Kh5 Ke7 8. Kh6 {8.h4 Kf7 9.Kh6 is just a different move order (a “minor dual” in a study).} Kf7 9. h4 {And again, 9.Ra6 Re8 10.h6 would just be a different move order.} Re8 {When you look at what happened in the game when Black allowed 10.Ra6, you just might suggest 9…Ra8 instead. But then both 9.Rb6 and 9.Rc6 are sufficient for a draw. There are some subtleties, though, in the case of 9...Rd8: 10.Ra6! f4, and here, in contrast to our game, 11.Ra7+? Kg8 12.Ra3 Rd6+ 13.Kh5 would be a blunder due to 13…Nd2! –+. Instead, White should opt for 11.Ra3! Rd6+ (11…Kg8 12.Kg6 =) 12.Kh7! = or 11.Ra4! which would lead us back to our game after 11…Re8 12.Ra7+!} 10. Ra6 $1 {It was just intuitive playing that led me move the rook as far as possible. In fact, the whole defence idea only works from the a-file.} (10. Rb6 $2 f4 11. Rb7+ (11. Rb3 Re6+ 12. Kh7 (12. Kh5 Rf6 $1 $19) 12... Nf6+ 13. Kh6 Re5 $19) 11... Kg8 12. Rb3 Re6+ 13. Kh5 Rf6 14. Kg4 Nd2 $19) 10... f4 11. Ra7+ $1 {It's crucial to first drive away the black king from the square where he protects the rook. Instead, immediate 11.Ra3? does not work here due to, for instance, 11…Re6+ 12.Kh5 (12.Kh7 Nf6+ 13.Kh6 Re5 –+) 12...Rf6 13.Kg4 Nc5 14.Rf3 Ne6 –+.} Kg8 12. Ra3 Re6+ 13. Kh5 Rf6 {Here, Black's plan does no longer work as the white king can attack the rook just in time. Other options include 13…Nf6+ 14.Kg5 Nd5 15.h5 =, 13...Nd2 14.Rd3 Ne4 15.Ra3 =, and 13...Kxg7 14.Kg4 Rf6 15.Rf3 =.} 14. Kg4 Nd2 (14... Nc5 15. Kg5 $1 Rf7 (15... Ne4+ 16. Kg4 $11) (15... Kxg7 16. Ra7+ Rf7 17. Rxf7+ Kxf7 18. Kxf4 $11) 16. Kh6 Rf6+ 17. Kg5 $11) (14... Kxg7 {or 14...Nd6} 15. Rf3 $11) (14... Ng3 15. Ra2 $1 Kxg7 16. Rf2 {see game}) 15. Kg5 $1 Rf7 (15... Kxg7 16. Ra7+ Rf7 17. Rxf7+ Kxf7 18. Kxf4 $11) (15... Ne4+ 16. Kg4 Nd2 17. Kg5 $11) 16. Kh6 Rf6+ 17. Kg5 Ne4+ 18. Kg4 Ng3 19. Ra2 $1 (19. Rf3 $2 Ne2 20. Rf2 Re6 $19) 19... Kxg7 20. Rf2 $11 {draw} 1/2-1/2

[Event "Simultaneous Exhibition"] [Site ""] [Date "1985"] [Round ""] [White "Viktor Korchnoi (ELO 2630)"] [Black "Volker Hergert"] [Result "1-0"] [Annotator "Hergert,Volker"] [SetUp "1"] [FEN "8/1p2K3/7P/P4kn1/8/8/5P2/8 w - - 0 48"] [PlyCount "11"] [SourceVersionDate "2026.02.24"] 
 {Nothing is more difficult than to win a won position. I guess, many would have failed to get the full point here… White must find three “only-moves” to achieve that goal…} 48. f4 $1 {Decisive decoy sacrifice to drive the white king away from the action. Instead, if 48.Kd7, then 48…Kg6 49.Kc7 Ne6+ 50.Kxb7 Nc5+ easily draws.} Kxf4 {48...Nh7 (as we have seen above, the knight belongs to the queenside) 49.Kd8 Kxf4 50.Kc7 +-} 49. Kd7 $1 {The only move that allows the king to go to the queenside without running into a check from the knight. If instead 49.Kd6, then 49…Nf7+ 50.Kc7 Nxh6 51.Kxb7 Nf5 52.a6 Nd6+ draws again.} Nf7 50. h7 $1 {White should not allow Nxh6, e.g. 50.Kc7 Nxh6 51.Kxb7 Nf5 = as in the side-line above.} Kf5 51. Kc7 Kg6 52. Kxb7 Kxh7 53. Kc7 $18 {1-0} (53. a6 $2 Nd6+ 54. Kc6 (54. Kc7 Nb5+ 55. Kc6 Na7+ $11) 54... Nc8 $11) 1-0

[Event "Saarbrücken"] [Site ""] [Date "1986"] [Round ""] [White "Herbert Bastian (ELO 2375)"] [Black "Volker Hergert"] [Result "1/2-1/2"] [Annotator "Hergert,Volker"] [SetUp "1"] [FEN "8/8/1p2k3/p3P3/4K3/8/1P1B1P2/r7 w - - 0 36"] [PlyCount "39"] [SourceVersionDate "2026.02.24"] 
 36. f4 $1 (36. Bc3 $2 a4 $1 37. f4 {too late} Rf1 38. Bb4 Rf2 39. Bc3 b5 40. Bd4 Re2+ 41. Kd3 Re1 42. Bc3 Rc1 43. Kd4 Rc2 44. Kd3 Rg2 45. Ke4 b4 46. Bxb4 (46. f5+ Kd7 47. e6+ Ke8 $19) 46... Re2+ $1 47. Kd3 Rxb2 $19) 36... Ra4+ {36...Rf1 37.b3! =, when White follows up with f5 and easily draws.} 37. b4 $1 axb4 38. f5+ Ke7 (38... Kf7 39. Kd5 Ra5+ 40. Kd6 $11) (38... Kd7 39. e6+ Kd6 40. Bf4+ Ke7 41. Bg5+ Kd6 42. Bf4+ $11) 39. Kd4 $1 {39.Kd3 b3 40.Kc3 same position as in the game; 39.Kd5? Ra5+ 40.Kc4 (40.Kc6 b3 -+) 40...Rc5+ 41.Kb3 (41.Kxb4 Rxe5 -+) After 42.f6+ Black should though avoid 42...Kf7? as 43.Ka4 or 42.Kc4 draws.) 41...Rxe5 42.Bxb4+ Kf7! –+. Now this is the correct move with 43.f6 answered by 43...Re3+ (–+).} b3+ 40. Kc3 Re4 41. Bg5+ $1 (41. Kxb3 $2 Rxe5 $19) 41... Ke8 42. e6 {Now good as the black king can no longer move to f6 as a reply.} Rg4 (42... Re5 43. Bh4 Rxf5 44. e7 $11) 43. Bf6 $1 (43. Bc1 $2 Rg3+ 44. Kc4 (44. Kb2 Rf3 $19) 44... Ke7 45. Ba3+ Kf6 46. e7 Re3 47. Kd4 b2 $1 48. Kxe3 (48. Bxb2 Rxe7 $19) 48... b1=Q 49. e8=Q Qe1+ 50. Kf4 Qxe8 $19) 43... Rg3+ (43... Rf4 44. e7 Rxf5 45. Bh4 $11) 44. Kb2 Rf3 45. e7 Rxf5 46. Bh4 {The same position as in the last sideline, but with the white king on b2 rather than c3.} Rh5 47. Bf6 {Despite a significantly favourable material situation, Black can make no progress.} Rh6 48. Bg5 Rg6 49. Bh4 Rg4 50. Bf6 Rf4 51. Bg5 Rb4 52. Bf6 Kf7 53. Bg5 Rg4 54. Bf6 Rf4 55. Bg5 $11 {draw} 1/2-1/2

[Event "Mainz"] [Site ""] [Date "1992"] [Round ""] [White "Volker Hergert"] [Black "Jens Beutel"] [Result "1/2-1/2"] [Annotator "Hergert,Volker"] [SetUp "1"] [FEN "k7/1p6/p7/8/5R1K/8/7P/2r5 w - - 0 0"] [PlyCount "29"] [SourceVersionDate "2026.02.25"] 
 {One of those typical rook endgames… this one proved to be far from purely technical!} 1. Kg5 b5 2. h4 Ka7 {Surprisingly, this is more precise than 2…Kb7.} 3. h5 Rc8 4. h6 {In the case of 2…Kb7 3.h5 Rc8, White would have a choice between 4.h6, 4.Rf6/Rg4 and 4.Kf6/Kg6! To understand this circumstance, let’s take the position from my game as a starting point. Here, all these candidate moves would lose to 4…a5 -+. As an example, we investigate 4.Kf6? a5 5.h6 a4 6.h7 a3 7.Rf3 b4 8.Rb3 a2 -+. With the white king on b7 instead of a7, 8..a2? would even lose due to Rxb3+ as the b-pawn is captured with check. After the correct moves 8…Ka6 9.Rxb4 it’s a draw. Something similar applies to 4.Kg6? a5 5.h6 a4 6.h7 a3. In my game, 7.Rf3 b4 8.Rb4 would now lose to 8…a2 -+ as well as 7.Rb4 a2 -+. But with the king on b7 instead of a7, we have 7.Rf3 b4 8.Rb3 a2? (8…Ka6 9.Rxb4 =) 9.Rxb4+ (+-) and 7.Rb4 a2 8.Rxb5+ Ka6 9.Rc5! =. That said, after the correct continuation 4.h6, both lines (2…Kb7 and 2…Ka7) lead to the same position. But with 2…Ka7, the move 4.h6 is in fact the only one that saves the game. ln the game, I was not aware of these subtleties.} a5 5. h7 a4 6. Rb4 $1 (6. Kg6 $2 {or 6.Kh6?} a3 $19) 6... Kb6 (6... a3 $2 7. Rxb5 $18) 7. Kh6 $1 {The g-file should be kept open for the rook!} Rh8 {Or 7...a3 8.Rb3 Ra8 (8…a2 9.Ra3 =) 9.Kg6/9.Kg5 a2 10.Ra3! Rxa3 11.h8Q a1Q 12.Qd8+/Qb8+ and White has a perpetual.} 8. Rg4 $1 (8. Kg7 $2 Rxh7+ $1 9. Kxh7 Ka5 $19 {or 9...Kc5 -+}) 8... a3 (8... Kc5 9. Rg5+ $1 (9. Rg8 $2 Rxh7+ 10. Kxh7 b4 11. Ra8 b3 12. Rxa4 b2 $19) 9... Kc4 10. Rg4+ Kc3 {or 10...Kb3} 11. Rg3+ $11) (8... Ka5 9. Rg8 Rxh7+ 10. Kxh7 $11) 9. Rg8 $1 {That’s why the white king had to leave the g-file.} Rxh7+ (9... a2 10. Rxh8 $11) 10. Kxh7 b4 11. Ra8 (11. Kg6 $2 a2 12. Ra8 b3 $19) 11... Kc5 12. Kg6 Kc4 13. Kf5 b3 (13... Kb3 14. Ke4 $11) 14. Ra4+ $1 (14. Rxa3 $2 b2 $19) 14... Kc3 (14... Kb5 15. Rxa3 $11) 15. Rxa3 $11 1/2-1/2

[Event "corr."] [Site ""] [Date "1998"] [Round ""] [White "Volker Hergert"] [Black "Markus Schachl (ELO 2200)"] [Result "1-0"] [Annotator "Hergert,Volker"] [SetUp "1"] [FEN "8/4R3/1Bp5/p7/2pp4/kn6/5P2/3K4 w - - 0 46"] [PlyCount "15"] [SourceVersionDate "2026.02.25"] 
 46. f4 $1 {Here, 46.Rd7 would be a mistake due to 46…c5 47.Kc2 (47.Rc7 Kb4 48.f4 c3 49.f5 d3 50.Bxc5+ Nxc5 51.f6 Ne4 =) 47…Kb4 48.f4 a4 49.Bd8 c3 50.Ra7 Na1+! 51.Kd3 Nb3 =. Instead, immediate advance makes the f-pawn the match winner. 46.Bxa5? Nxa5 47.Ra7 Kb4 does not yet work. After 48.f4 c3 49.f5 Black’s knight moves to b3 or c4 and secures a draw. If instead 49.Rxa5? Kxa5 50.f5, Black is just in time to promote his c-pawn: 50…Kb4 51.f6 Kb3 52.f7 (or 52.Kc1 d3 -+) 52…Kb2 -+. These are tablebase wins for Black, though, and would be hard to prove over the board. But in a correspondence game it would be no problem to convert.} a4 (46... d3 {and now the bishop sacrifice works perfectly well being one tempo ahead compared to 46.Bxa5:} 47. Bxa5 $1 Nxa5 48. Ra7 Kb4 49. Rxa5 $1 Kxa5 50. f5 c3 51. f6 Kb4 52. f7 Kb3 53. f8=Q $18) 47. Rd7 $1 ({And now this other candidate from one move earlier works!} 47. f5 $2 d3 48. f6 c3 49. Be3 c2+ 50. Ke1 Kb2 $11) 47... c5 (47... Kb4 48. f5 $18) 48. Rc7 $1 (48. Kc2 $2 Kb4 49. Bd8 c3 50. Ra7 Na1+ $1 51. Kd3 Nb3 $11 {(as already shown above)}) 48... Kb4 (48... c3 49. Bxc5+ Kb2 50. Bd6 {White could also opt for another square on the a3-f8 diagonal. After 50…a3 it’s no difference.} a3 51. Bxa3+ Kxa3 52. f5 $18 {or 52.Kc2 +-}) 49. f5 {Here, White wins as I’m one tempo ahead of the line shown above.} Kb5 (49... c3 50. f6 d3 51. Bxc5+ $1 Nxc5 52. Rxc5 $1 Kxc5 53. f7 $18) (49... a3 50. f6 a2 51. Ra7 $18) 50. Bxc5 $1 Nxc5 51. f6 a3 52. Rxc5+ $1 Kxc5 53. f7 1-0

[Event "Königstein"] [Site ""] [Date "2001"] [Round ""] [White "Volker Hergert"] [Black "Horst-Hermann Stein"] [Result "1-0"] [Annotator "Hergert,Volker"] [SetUp "1"] [FEN "5rk1/6pp/8/4Qp1q/8/6P1/4RPP1/6K1 w - - 0 0"] [PlyCount "23"] [SourceVersionDate "2026.02.25"] 
 {The position is fully balanced, and I already considered offering a draw, when I noticed a chance to complicate things.} 1. g4 {To be fair: when I made the move, I was not sure whether the pawn was really poisoned. And given that this was an informal friendship game I would probably also have captured it just to find out whether the sacrifice was correct.} Qxg4 $2 {With perfect hindsight, Black should just withdraw his queen and after, for instance, 1…Qf7 the situation remains as balanced as it was before.} 2. Qd5+ $1 (2. Qe6+ $2 Rf7 $1 $11 {the e-file should be open for the rook}) 2... Kh8 3. Qf7 $1 {The idea of the pawn sac becomes visible: Black’s queen was deviated from protecting the critical squares f7 and e8.} Rg8 (3... Qb4 4. Re8 Qb1+ 5. Kh2 $18) 4. Re8 Qd1+ 5. Kh2 Qd6+ 6. Kh3 $1 {6.g3? leads to a perpetual on h6 and c1, because White’s king should not go to g2 (Qc6+, -+).} Qh6+ 7. Kg3 Qg6+ {Black can’t achieve a perpetual and that’s why he trades queens. The alternative would have been 7…Qg5+ 8.Kf3 Qg4+ 9.Ke3 f4+ (9…Qg5+ 10.Kd3 +-) 10.Kd2 +-.} 8. Qxg6 hxg6 9. Rxg8+ Kxg8 10. Kf4 $1 {White is a pawn down, but after this king advance, the pure pawn endgame is easily won for White. It just needs precise timing with the f-pawn. By the way, 10.Kh4? would have been a blunder due to 10…f4! 11.Kg5 (11.Kg4 Kf7 12.Kxf4 Kf6 13.g4 g5+ 14.Ke4 Ke6 =) 11…f3! (11…Kf7? 12.f3! +-) 12.gxf3 (12.g3 Kf7 13.g4 Ke6 14.Kxg6 Ke5 =) 12…Kf7 13.f4 Ke6 14.Kxg6 Kd5 15.f3 Ke6 16.f5+ Ke5 17.Kg5 Kd5 =.} Kf7 (10... Kf8 11. Kg5 Kf7 12. f4 $1 $18) 11. Kg5 f4 12. f3 $1 {mutual zugzwang} (12. Kxf4 $2 Kf6 13. g4 g5+ 14. Ke4 Ke6 $11) 1-0

[Event "London"] [Site ""] [Date "2014"] [Round ""] [White "Hergert"] [Black "Milner"] [Result "1-0"] [Annotator "Hergert,Volker"] [SetUp "1"] [FEN "8/1R6/8/P2k4/n2pp3/B1p5/2K5/8 b - - 0 50"] [PlyCount "26"] [SourceVersionDate "2026.02.25"] 
 50... d3+ 51. Kb3 $1 ({It was a quick decision to direct the king on the most active square, and probably it was a good intuition. The weaker alternative was} 51. Kd1 $2 e3 {and here it is fascinating to see how Black achieves either a perpetual or with White’s most stubborn attempts a stalemate in a forced way:} 52. Rc7 (52. Bc1 d2 53. Ba3 Ke4 54. a6 Kf3 55. Rf7+ Kg3 56. a7 Nb2+ 57. Bxb2 cxb2 58. Rb7 Kf2 59. Rxb2 e2+ 60. Kxd2 e1=Q+ 61. Kd3+ Kg1 62. a8=Q Qd1+ (62... Qf1+ 63. Re2 Qf3+ (63... Qd1+ 64. Rd2 Qf3+ 65. Qxf3 {stalemate}) 64. Qxf3 {stalemate}) 63. Rd2 Qf3+ 64. Qxf3 {stalemate}) 52... Ke4 53. a6 Kf3 54. Rf7+ Kg3 55. a7 Nb2+ 56. Bxb2 cxb2 57. Rb7 Kf2 58. Rxb2+ e2+ 59. Kd2 e1=Q+ 60. Kxd3+ Kg1 61. a8=Q Qd1+ (61... Qf1+ 62. Re2 Qf3+ (62... Qd1+ 63. Rd2 Qf3+ 64. Qxf3 {stalemate}) 63. Qxf3 {stalemate}) 62. Rd2 Qf3+ 63. Qxf3 {stalemate}) 51... e3 52. Rd7+ Ke4 53. a6 {The knight is taboo. White must hurry up getting a new queen.} e2 (53... d2 54. a7 e2 (54... Nb6 55. Kxc3 $18) 55. a8=Q+ Ke3 56. Re7+ Kf2 57. Rf7+ Ke1 58. Qh1#) 54. a7 Nb6 ({Black must prevent White from getting his queen with check:} 54... e1=Q 55. a8=Q+ Ke3 56. Re7+ Kd2 57. Qg2+ Kd1 58. Qf3+ Kd2 59. Bc1+ Qxc1 60. Qf2+ Kd1 61. Re1#) 55. Re7+ Kd4 56. Bb4 $1 (56. Bc5+ $2 Kxc5 57. Kxc3 Kd6 $1 $11) 56... d2 57. Bxc3+ Kd3 {Now White in quite a brutal way sacrifices both pieces to force the emergence of a new queen.} 58. Bxd2 Kxd2 59. Rxe2+ (59. Kb4 $2 Nd5+ $11) 59... Kxe2 60. Kb4 Kd3 61. Kb5 Nd5 {The very last trap.} 62. Kc5 (62. a8=Q $2 Nc7+ $11) 62... Nc7 63. Kc6 1-0

[Event "Stockholm"] [Site ""] [Date "2016"] [Round ""] [White "Lund"] [Black "Hergert"] [Result "1-0"] [Annotator "Hergert,Volker"] [SetUp "1"] [FEN "6R1/2p5/1p4P1/pPkPp3/P1P3r1/2K5/8/8 w - - 0 51"] [PlyCount "21"] [SourceVersionDate "2026.02.26"] 
 51. g7 {Threatens 52.d6 +-.} Rg3+ (51... Rxc4+ 52. Kd2 Rg4 53. d6 $1 $18) 52. Kd2 $1 (52. Kc2 $2 Kxc4 53. d6 Rg2+ 54. Kd1 Kd3 $11) 52... Kd4 (52... Kxc4 53. d6 $1 $18) 53. c5 $1 {What a tragedy! Black had to renounce on capturing this pawn and now it breaks through! Instead, 53.Ke2? fails to 53…Rg2+ 54.Kf3 Rg5 = and 53.d6? cxd6 54.c5 would let Black escape to 54…Rg2+ 55.Ke1 Ke3 56.Kf1 Rf2+ 57.Kg1 Rf7=.} e4 {The pawn is untouchable: 53...bxc5 54.Ke2! +- resp. 53...Kxc5 54.d6! +-, while 53...Rg2+ 54.Ke1 Ke3 (54...e4 55.c6 +- ) 55.Kf1 Rf2+ 56.Kg1 Rf7 57.c6! e4 58.Kh2 leads to the main line.} 54. c6 $1 (54. d6 $2 cxd6 55. cxb6 e3+ 56. Ke2 Rg2+ 57. Kf3 Rf2+ 58. Kg4 Rf7 $11) 54... Rg2+ (54... e3+ 55. Ke2 Rg2+ 56. Kf3 e2 (56... Rf2+ 57. Kg4 $18) 57. Re8 Rxg7 58. d6 cxd6 59. Kxe2 $1 $18) 55. Ke1 Ke3 56. Kf1 Rf2+ 57. Kg1 Rf7 58. Kh2 $1 Kd4 {58...Kd2 and 58...Ke2 are refuted in the same way.} 59. Rc8 $1 Rxg7 60. Rxc7 $1 Rxc7 61. d6 1-0